With the development of communications technologies, a multipoint video conference in daily life and work becomes an effective platform for people to communicate in different places, and is more widely applied in various application fields and scenarios.
In a conventional multipoint video conference, generally a multipoint control unit (Multipoint Control Unit, MCU for short) is adopted to control screen display of each site participating in the conference, and the MCU controls the screen display according to a preset display policy. Specifically, a conference controller needs to separately formulate a display policy for the site according to the characteristic of each site, for example, obtain a quantity of display screens of a certain site in advance, and formulates a display policy specific to the site according to the obtained number of display screens of the site, that is, settings of display content and a display manner of each display screen in the site, and then convert the display policy into a configuration file and import it to an MCU. When a video conference is ongoing, the MCU delivers to the site, according to the formulated display policy, received code stream information sent to the MCU by each sitesite, so as to display an image corresponding to the preset policy on a display screen of the site. For example, three sites in Beijing, Shanghai and Nanjing need to jointly hold a video conference, where a quantity of display screens in Beijing branch site is 3, a quantity of display screens in Shanghai branch site is 2, and a quantity of display screens in Nanjing branch site is 1. After obtaining a quantity of display screens in the three participating sites in Beijing, Shanghai and Nanjing, an MCU separately formulates a corresponding display policy for the three sites according to the obtained number of display screens. For example, in the beginning phase, the display policy of Beijing branch site is: the left display screen displays, in a split screen mode, an image included in a code stream of the Shanghai branch site, the middle display screen displays local Power Point slides, and the right display screen displays an image included in a code stream of the Nanjing branch site. These display policies are formulated into a configuration file to be imported to an MCU of a video conference system. When a conference is ongoing, in the beginning phase of the conference, the display screens of the Beijing branch site will display images on local screens according to the formulated display policy.
The code stream processing method of the conventional multipoint video conference has the following defects: the display policy is separately formulated according to the characteristic of each site, site policies may vary according to sites with different characteristics, and when formulating a display policy of a site, a quantity of display screens of each site participating in the video conference must be obtained in advance, and the display policy for each site is formulated according to the obtained number of display screens. When a quantity of screens of the sites participating in the conference changes, a user needs to reformulate a display policy; therefore, the adaptability of the display policy is poor in the multipoint video conference, and it is inconvenient for the user to use.